Improving Ki67 assessment concordance by the use of an artificial intelligence-empowered microscope: a multi-institutional ring study

Artificial intelligence entering the pathology arena in oncology: current applications and future perspectives

BACKGROUND

Artificial intelligence (AI) is rapidly transforming the fields of pathology and oncology, offering novel opportunities for advancing diagnosis, prognosis, and treatment of cancer.

METHODS

Through a systematic review-based approach, the representatives from the European Society for Medical Oncology (ESMO) Precision Oncology Working Group (POWG) and international experts identified studies in pathology and oncology that applied AI-based algorithms for tumour diagnosis, molecular biomarker detection, and cancer prognosis assessment. These findings were synthesised to provide a comprehensive overview of current AI applications and future directions in cancer pathology.

RESULTS

The integration of AI tools in digital pathology is markedly improving the accuracy and efficiency of image analysis, allowing for automated tumour detection and classification, identification of prognostic molecular biomarkers, and prediction of treatment response and patient outcomes. Several barriers for the adoption of AI in clinical workflows, such as data availability, explainability, and regulatory considerations, still persist. There are currently no prognostic or predictive AI-based biomarkers supported by level IA or IB evidence. The ongoing advancements in AI algorithms, particularly foundation models, generalist models and transformer-based deep learning, offer immense promise for the future of cancer research and care. AI is also facilitating the integration of multi-omics data, leading to more precise patient stratification and personalised treatment strategies.

CONCLUSIONS

The application of AI in pathology is poised to not only enhance the accuracy and efficiency of cancer diagnosis and prognosis but also facilitate the development of personalised treatment strategies. Although barriers to implementation remain, ongoing research and development in this field coupled with addressing ethical and regulatory considerations will likely lead to a future where AI plays an integral role in cancer management and precision medicine. The continued evolution and adoption of AI in pathology and oncology are anticipated to reshape the landscape of cancer care, heralding a new era of precision medicine and improved patient outcomes.

Evaluating the Impact of a Ki-67 Decision Support Algorithm on Pathology Residents' Scoring Accuracy

Ki-67 scoring is of essential importance in the evaluation of breast cancer. We evaluated a Ki-67 algorithm as a decision support tool to improve accuracy for pathology residents. We retrospectively evaluated Ki-67 scores on whole slide images (WSI) obtained from 156 consecutive breast cancer patients. Two senior pathologists determined the 2.1 mm2 hotspot to be evaluated. Ki-67 scores from senior pathologists were compared with results generated by the algorithm, results from 10 pathology residents, and results from pathology residents with the assistance of the algorithm. In addition to numerical results from the algorithm, residents were also presented with a visual representation of nuclei that were counted and excluded. Statistical analysis was performed using Wilcoxon and intra-class correlation (ICC) tests. The mean Ki-67 scores from senior pathologists and the algorithm were 23 ± 18 and 24 ± 18, respectively (ICC, 0.98). Ki-67 scores from the residents were 19 ± 16 and 22 ± 16, without and with input from the algorithm, respectively. With input from the algorithm, residents' scores were significantly closer to those obtained by senior pathologists (p = 0.008). Residents modified their scores in 53.8% of the cases where 74% of the better scores were characterized by an increase in the original scores. The results obtained by the Ki-67 algorithm were highly correlated with those assessed by senior pathologists. We demonstrated that the algorithm may serve as a decision support tool for residents to align their results with those of senior pathologists.

Histopathologic Differential Diagnosis and Estrogen Receptor/Progesterone Receptor Immunohistochemical Evaluation of Breast Carcinoma Using a Deep Learning-Based Artificial Intelligence Architecture

In breast carcinoma, invasive ductal carcinoma (IDC) is the most common histopathologic subtype, and ductal carcinoma in situ (DCIS) is a precursor of IDC. These two often occur concomitantly. The immunohistochemical staining of estrogen receptor (ER)/progesterone receptor (PR) in IDC/DCIS on histopathologic whole slide images (WSIs) can predict the prognosis of patients. Artificial intelligence (AI) technology has the potential to substantially reduce the interobserver variability among pathologists reading WSIs. Herein, IDC/DCIS detection was conducted by a deep learning approach, including faster region-based convolutional neural network (Faster R-CNN), RetinaNet, single-shot multibox detector 300 (SSD300), you only look once (YOLO) v3, YOLOv5, YOLOv7, YOLOv8, and Swin transformer. Their performance was estimated by mean average precision (mAP) values. Cell recognition and counting were performed using AI technology to evaluate the intensity and proportion of ER/PR-immunostained cancer cells in IDC/DCIS. A three-round ring study (RS) was conducted to assess WSIs. A database for modelling the underlying probability distribution of a data set with labels was established. YOLOv8 exhibited the highest detection performance with an mAP at 0.5 of 0.944 and an mAP at 0.5 to 0.95 of 0.790. With the assistance of YOLOv8, the scoring concordance across all pathologists was boosted to excellent in RS3 (0.970) from moderate in RS1 (0.724) and good in RS2 (0.812). Deep learning detection can be applied in the clinicopathologic field. Herein, a novel AI architecture and well-organized data set were developed to facilitate the histopathologic diagnosis of IDC/DCIS and immunostaining scoring of ER/PR.

Correlation analysis of Ki67 changes with survival outcomes in breast cancer before and after neoadjuvant therapy based on residual cancer Burden grade

PURPOSE

This study aims to investigate the change of Ki67 value pre- and post-neoadjuvant therapy (NAT) and evaluate its potential value in predicting survival outcomes in different molecular subtypes of breast cancer.

METHODS

A total of 257 breast cancer patients who underwent NAT at Renmin Hospital of Wuhan University from July 2019 to Sep 2023 were included in this study. The Ki67 index of the patients was re-interpreted by two attending physicians, and the changes of Ki67 value pre- and post-NAT were compared. Chi-square test (χ2) and logistic regression were conducted to examine the correlation between various characteristics and the efficacy of NAT. Disease-free survival (DFS) was calculated using the Kaplan-Meier curve and compared using the log-rank test.

RESULTS

Patients with higher histological grade, negative expression of estrogen receptor (ER) or progesterone receptor (PR), positive expression of human epidermal growth receptor 2 (HER2), higher pretreatment Ki67 index, absence of lymph node metastasis, and those with HER2 positive and triple-negative breast cancer were associated with improved efficacy of NAT. Our study identified that the optimal cut-off value for the changes in Ki67 index pre- and post-NAT related to the effectiveness of NAT was "-88.19 %" in whole chort, which was related to the aforementioned clinical characteristics. Besides, the optimal cut-off values for the luminal, HER2-enriched and triple-negative subtypes were "-91.83 %", "-46.12 %" and "-81.67 %", respectively. Survival analysis demonstrated that the changes in Ki67 value were significantly associated with DFS in the HER2-enriched and triple-negative subtype, but not in the luminal subtype.

CONCLUSIONS

Preoperative clinicopathological features and changes in Ki67 value pre-and post-NAT can contribute to providing patients with a more accurate prognosis.

Artificial intelligence enhances whole-slide interpretation of PD-L1 CPS in triple-negative breast cancer: A multi-institutional ring study

BACKGROUND AND AIMS

Evaluation of the programmed cell death ligand-1 (PD-L1) combined positive score (CPS) is vital to predict the efficacy of the immunotherapy in triple-negative breast cancer (TNBC), but pathologists show substantial variability in the consistency and accuracy of the interpretation. It is of great importance to establish an objective and effective method which is highly repeatable.

METHODS

We proposed a model in a deep learning-based framework, which at the patch level incorporated cell analysis and tissue region analysis, followed by the whole-slide level fusion of patch results. Three rounds of ring studies (RSs) were conducted. Twenty-one pathologists of different levels from four institutions evaluated the PD-L1 CPS in TNBC specimens as continuous scores by visual assessment and our artificial intelligence (AI)-assisted method.

RESULTS

In the visual assessment, the interpretation results of PD-L1 (Dako 22C3) CPS by different levels of pathologists have significant differences and showed weak consistency. Using AI-assisted interpretation, there were no significant differences between all pathologists (P = 0.43), and the intraclass correlation coefficient (ICC) value was increased from 0.618 [95% confidence interval (CI) = 0.524-0.719] to 0.931 (95% CI = 0.902-0.955). The accuracy of interpretation result is further improved to 0.919 (95% CI = 0.886-0.947). Acceptance of AI results by junior pathologists was the highest among all levels, and 80% of the AI results were accepted overall.

CONCLUSION

With the help of the AI-assisted diagnostic method, different levels of pathologists achieved excellent consistency and repeatability in the interpretation of PD-L1 (Dako 22C3) CPS. Our AI-assisted diagnostic approach was proved to strengthen the consistency and repeatability in clinical practice.

Gastroenteropancreatic neuroendocrine neoplasms: current development, challenges, and clinical perspectives

Neuroendocrine neoplasms (NENs) are highly heterogeneous and potentially malignant tumors arising from secretory cells of the neuroendocrine system. Gastroenteropancreatic neuroendocrine neoplasms (GEP-NENs) are the most common subtype of NENs. Historically, GEP-NENs have been regarded as infrequent and slow-growing malignancies; however, recent data have demonstrated that the worldwide prevalence and incidence of GEP-NENs have increased exponentially over the last three decades. In addition, an increasing number of studies have proven that GEP-NENs result in a limited life expectancy. These findings suggested that the natural biology of GEP-NENs is more aggressive than commonly assumed. Therefore, there is an urgent need for advanced researches focusing on the diagnosis and management of patients with GEP-NENs. In this review, we have summarized the limitations and recent advancements in our comprehension of the epidemiology, clinical presentations, pathology, molecular biology, diagnosis, and treatment of GEP-NETs to identify factors contributing to delays in diagnosis and timely treatment of these patients.

Association Between Ki-67 Proliferative Index and Oncotype-Dx Recurrence Score in Hormone Receptor-Positive, HER2-Negative Early Breast Cancers. A Systematic Review of the Literature

Background

Oncotype-Dx (ODx) is a 21-gene assay used as a prognostic and predictive tool for hormone receptor (HR)-positive and human epidermal growth factor receptor 2 (HER2)-negative, node-negative, or 1 to 3 lymph node-positive early breast cancers (EBCs). The cost of the test, which is not available in low-middle income countries (LMICs), is not within the means of most individuals. The Ki-67 index is a marker of tumor proliferation that is cost-effective and easily performed and has been substituted in many cases to obtain prognostic information.

Objective

We aimed to identify the correlation between the ODx recurrence score (RS) and the Ki-67 index in HR-positive EBCs and to determine whether Ki-67, like the ODx, can help facilitate clinical decision-making.

Design

Systematic review correlating Ki-67 index and ODx in HR-positive and HER2-negative EBCs as per Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

Data sources and methods

We searched different databases between January 2010 and May 2023 and included retrospective/prospective cohorts, clinical trials, case-control, and cross-sectional studies involving HR-positive and HER2-negative EBCs correlating the Ki-67 index and ODx RS categories.

Results

Of the 18 studies included, 16 indicated a positive or weakly positive correlation between ODx and the Ki-67 index. The combined P value of the included studies is <0.05 (P = .000), which shows a statistical significance between the 2. Our review also discusses the potential of machine learning and artificial intelligence (AI) in Ki-67 assessment, offering a cost-effective and reproducible alternative.

Conclusion

Even although there are limitations, studies indicate a favorable association between ODx and the Ki-67 index in specific situations. This implies that Ki-67 can offer important predictive details, especially regarding the likelihood of relapse in HR-positive EBC. This is particularly significant in LMICs where financial constraints often hinder the availability of costly diagnostic tests.

LYSTO: The Lymphocyte Assessment Hackathon and Benchmark Dataset

We introduce LYSTO, the Lymphocyte Assessment Hackathon, which was held in conjunction with the MICCAI 2019 Conference in Shenzhen (China). The competition required participants to automatically assess the number of lymphocytes, in particular T-cells, in images of colon, breast, and prostate cancer stained with CD3 and CD8 immunohistochemistry. Differently from other challenges setup in medical image analysis, LYSTO participants were solely given a few hours to address this problem. In this paper, we describe the goal and the multi-phase organization of the hackathon; we describe the proposed methods and the on-site results. Additionally, we present post-competition results where we show how the presented methods perform on an independent set of lung cancer slides, which was not part of the initial competition, as well as a comparison on lymphocyte assessment between presented methods and a panel of pathologists. We show that some of the participants were capable to achieve pathologist-level performance at lymphocyte assessment. After the hackathon, LYSTO was left as a lightweight plug-and-play benchmark dataset on grand-challenge website, together with an automatic evaluation platform.

AI improves accuracy, agreement and efficiency of pathologists for Ki67 assessments in breast cancer

The Ki-67 proliferation index (PI) guides treatment decisions in breast cancer but suffers from poor inter-rater reproducibility. Although AI tools have been designed for Ki-67 assessment, their impact on pathologists' work remains understudied. 90 international pathologists were recruited to assess the Ki-67 PI of ten breast cancer tissue microarrays with and without AI. Accuracy, agreement, and turnaround time with and without AI were compared. Pathologists' perspectives on AI were collected. Using AI led to a significant decrease in PI error (2.1% with AI vs. 5.9% without AI, p < 0.001), better inter-rater agreement (ICC: 0.70 vs. 0.92; Krippendorff's α: 0.63 vs. 0.89; Fleiss' Kappa: 0.40 vs. 0.86), and an 11.9% overall median reduction in turnaround time. Most pathologists (84%) found the AI reliable. For Ki-67 assessments, 76% of respondents believed AI enhances accuracy, 82% said it improves consistency, and 83% trust it will improve efficiency. This study highlights AI's potential to standardize Ki-67 scoring, especially between 5 and 30% PI-a range with low PI agreement. This could pave the way for a universally accepted PI score to guide treatment decisions, emphasizing the promising role of AI integration into pathologist workflows.

AI-Powered Segmentation of Invasive Carcinoma Regions in Breast Cancer Immunohistochemical Whole-Slide Images

AIMS

The automation of quantitative evaluation for breast immunohistochemistry (IHC) plays a crucial role in reducing the workload of pathologists and enhancing the objectivity of diagnoses. However, current methods face challenges in achieving fully automated immunohistochemistry quantification due to the complexity of segmenting the tumor area into distinct ductal carcinoma in situ (DCIS) and invasive carcinoma (IC) regions. Moreover, the quantitative analysis of immunohistochemistry requires a specific focus on invasive carcinoma regions.

METHODS AND RESULTS

In this study, we propose an innovative approach to automatically identify invasive carcinoma regions in breast cancer immunohistochemistry whole-slide images (WSIs). Our method leverages a neural network that combines multi-scale morphological features with boundary features, enabling precise segmentation of invasive carcinoma regions without the need for additional H&E and P63 staining slides. In addition, we introduced an advanced semi-supervised learning algorithm, allowing efficient training of the model using unlabeled data. To evaluate the effectiveness of our approach, we constructed a dataset consisting of 618 IHC-stained WSIs from 170 cases, including four types of staining (ER, PR, HER2, and Ki-67). Notably, the model demonstrated an impressive intersection over union (IoU) score exceeding 80% on the test set. Furthermore, to ascertain the practical utility of our model in IHC quantitative evaluation, we constructed a fully automated Ki-67 scoring system based on the model's predictions. Comparative experiments convincingly demonstrated that our system exhibited high consistency with the scores given by experienced pathologists.

CONCLUSIONS

Our developed model excels in accurately distinguishing between DCIS and invasive carcinoma regions in breast cancer immunohistochemistry WSIs. This method paves the way for a clinically available, fully automated immunohistochemistry quantitative scoring system.

Biological insights and novel biomarker discovery through deep learning approaches in breast cancer histopathology

Breast cancer remains a highly prevalent disease with considerable inter- and intra-tumoral heterogeneity complicating prognostication and treatment decisions. The utilization and depth of genomic, transcriptomic and proteomic data for cancer has exploded over recent times and the addition of spatial context to this information, by understanding the correlating morphologic and spatial patterns of cells in tissue samples, has created an exciting frontier of research, histo-genomics. At the same time, deep learning (DL), a class of machine learning algorithms employing artificial neural networks, has rapidly progressed in the last decade with a confluence of technical developments - including the advent of modern graphic processing units (GPU), allowing efficient implementation of increasingly complex architectures at scale; advances in the theoretical and practical design of network architectures; and access to larger datasets for training - all leading to sweeping advances in image classification and object detection. In this review, we examine recent developments in the application of DL in breast cancer histology with particular emphasis of those producing biologic insights or novel biomarkers, spanning the extraction of genomic information to the use of stroma to predict cancer recurrence, with the aim of suggesting avenues for further advancing this exciting field.

A scoping review of deep learning in cancer nursing combined with augmented reality: The era of intelligent nursing is coming

Artificial intelligence has been developing greatly in the field of medicine. As a new research hotspot of artificial intelligence, deep learning (DL) has been widely applied in the fields of cancer risk assessment, symptom recognition, and cancer detection. Therefore, nursing care issues in terms of consuming time and energy, lower accuracy, and lower efficiency can be solved with applying DL in caring cancer patients. In addition, augmented reality (AR) has great navigation potential through combining computer-generated virtual elements with the real world. Thus, DL ​+ ​AR may facilitate patients with cancer to possess a brand-new model of nursing care that is more intelligent, mobile, and adapted to the information age, compared to traditional nursing. With the advent of the era of intelligent nursing, future nursing models can not only learn from the DL ​+ ​AR model to meet the needs of patients with cancer but also reduce nursing workload, save healthcare resources, and improve work efficiency, the quality of nursing care, as well as the quality of life for cancer patients.

Deep learning-based single-shot autofocus method for digital microscopy

Digital pathology is being transformed by artificial intelligence (AI)-based pathological diagnosis. One major challenge for correct AI diagnoses is to ensure the focus quality of captured images. Here, we propose a deep learning-based single-shot autofocus method for microscopy. We use a modified MobileNetV3, a lightweight network, to predict the defocus distance with a single-shot microscopy image acquired at an arbitrary image plane without secondary camera or additional optics. The defocus prediction takes only 9 ms with a focusing error of only ∼1/15 depth of field. We also provide implementation examples for the augmented reality microscope and the whole slide imaging (WSI) system. Our proposed technique can perform real-time and accurate autofocus which will not only support pathologists in their daily work, but also provide potential applications in the life sciences, material research, and industrial automatic detection.

The additional value of ONEST (Observers Needed to Evaluate Subjective Tests) in assessing reproducibility of oestrogen receptor, progesterone receptor, and Ki67 classification in breast cancer

The reproducibility of assessing potential biomarkers is crucial for their implementation. ONEST (Observers Needed to Evaluate Subjective Tests) has been recently introduced as a new additive evaluation method for the assessment of reliability, by demonstrating how the number of observers impact on interobserver agreement. Oestrogen receptor (ER), progesterone receptor (PR), and Ki67 proliferation marker immunohistochemical stainings were assessed on 50 core needle biopsy and 50 excision samples from breast cancers by 9 pathologists according to daily practice. ER and PR statuses based on the percentages of stained nuclei were the most consistently assessed parameters (intraclass correlation coefficients, ICC 0.918-0.996), whereas Ki67 with 5 different theoretical or St Gallen Consensus Conference-proposed cut-off values demonstrated moderate to good reproducibility (ICC: 0.625-0.760). ONEST highlighted that consistent tests like ER and PR assessment needed only 2 or 3 observers for optimal evaluation of reproducibility, and the width between plots of the best and worst overall percent agreement values for 100 randomly selected permutations of observers was narrow. In contrast, with less consistently evaluated tests of Ki67 categorization, ONEST suggested at least 5 observers required for more trustful assessment of reliability, and the bandwidth of the best and worst plots was wider (up to 34% difference between two observers). ONEST has additional value to traditional calculations of the interobserver agreement by not only highlighting the number of observers needed to trustfully evaluate reproducibility but also by highlighting the rate of agreement with an increasing number of observers and disagreement between the better and worse ratings.

H2AFZ: A Novel Prognostic Marker in Canine Melanoma and a Predictive Marker for Resistance to CDK4/6 Inhibitor Treatment

Uncontrolled proliferation is a key feature of tumor progression and malignancy. This suggests that cell-cycle related factors could be exploited as cancer biomarkers and that pathways specifically involved in the cell cycle, such as the Rb-E2F pathway, could be targeted as an effective anti-tumor therapy. We investigated 34 formalin-fixed paraffin-embedded (FFPE) tissue samples of canine cutaneous melanocytoma, cutaneous melanoma, and oral melanoma. Corresponding clinical follow-up data were used to determine the prognostic value of the mRNA expression levels of several cell cycle regulated E2F target genes (E2F1, DHFR, CDC6, ATAD2, MCM2, H2AFZ, GINS2, and survivin/BIRC5). Moreover, using four canine melanoma cell lines, we explored the possibility of blocking the Rb-E2F pathway by using a CDK4/6 inhibitor (Palbociclib) as a potential anti-cancer therapy. We investigated the expression levels of the same E2F target gene transcripts before and after treatment to determine the potential utility of these molecules as predictive markers. The E2F target gene H2AFZ was expressed in 91.43% of the primary tumors and H2AFZ expression was significantly higher in cases with unfavorable clinical outcome. Among the other tested genes, survivin/BIRC5 showed as well-promising results as a prognostic marker in canine melanoma. Three of the four tested melanoma cell lines were sensitive to the CDK4/6 inhibitor. The resistant cell line displayed higher expression levels of H2AFZ before treatment compared to the CDK4/6 inhibitor-sensitive cell lines. The present results suggest that CDK4/6 inhibitors could potentially be used as a new anti-cancer treatment for canine melanoma and that H2AFZ could serve as a prognostic and predictive marker for patient selection.